Direkt zum Inhalt

SmartLab Studio - Artificial Intelligence Powered

SmartLab Studio II splash screen

SmartLab Studio II

Artificial Intelligence Powered

Works with SmartLab Studio II version 5.0 with the relevant plugin (Powder Plugin, XRR Plugin or Cluster Analysis option)

SmartLab Studio II is a Windows®-based software suite developed for the flagship Rigaku SmartLab X-ray diffractometer that integrates user privileges, measurements, analyses, data visualization and reporting. Newly available for the MiniFlex, the modular (plugin) architecture of this software delivers state-of-the-art interoperability between the functional components.

Explore the new AI functionality available.

How does AI help me?

This module can improve your productivity when you often analyze similar samples but have difficulty identifying minor phases such as impurities, foreign materials, etc.

A trained network can perform phase identification more accurately than conventional search/match algorithms without any operator involvement or judgment..

How does AI learn to identify phases?

This module uses a neural network-based deep-learning algorithm. A neural network can be trained and learn how typical main and minor phases and hundreds of possible minor phases can appear in X-ray diffraction (XRD) patterns using simulated data.

AI schematic

How does a neural network learn?

The neural network is modeled loosely on the human brain. The network architecture has input, hidden, and output layers with a number of “neurons.” The neurons are connected between the layers, and random weights and biases are assigned to these connections to set their “importance.” During training, the network processes input data (a simulated XRD pattern) and predicts the output (a set of phases) using the initial weights and biases. Then, the prediction is compared to the ground truth. Depending on how close the output is to the ground truth, a “reward” or “penalty” is propagated through the network as “feedback.” By repeating this process, the network refines the weights and biases of the connections, which is equivalent to “learning” the relationship between the input and output.

Example 1

We analyzed the same XRD pattern using a traditional search/match algorithm and the AI-powered phase identification and assessed their accuracy as the percentage of correctly identified phases. (If three out of four are correct, the accuracy is 3/4 = 75%, for example.)

When applied to a relatively simple mixture of five mineral phases, the AI-powered phase identification improved the average accuracy of phase identification without operator involvement from 79% to 95%. The result of AI-powered phase identification is shown below.

Example 1 graphic
Simple sample: Minerals (5 phases)

 

Even with a more complicated mixture of five cement phases, the improvement was from 45% to 80%. The result of AI-powered phase identification is shown below.

Example 1 graphic
Complex sample: Cement (5 phases)

Example 2

Another way to assess the correctness of phase identification results is to run a Rietveld analysis to see if the identified phases can successfully reproduce the measured profile and how accurately they are quantified.

In this example, the AI-powered phase identification was performed on the measured profile of NIST2686 (clinker cement standard sample), and their quantities were refined using the Rietveld method.

Example 1 graphic
NIST2686 Rietveld analysis results)

 
Weight fraction, wt%
  Conventional
search/match
AI-powered
phase ID
NIST certified
values
C3S 44.5 ± 1.2 57.5 ± 0.8 58.6 ± 4.0
C2S 35.1 ± 1.2 24.6 ± 0.8 23.3 ± 2.8
C3A 0.2 ± 1.4 0.3 ± 0.6 2.3 ± 2.1
C4AF 14.0 ± 0.8 13.1 ± 0.6 14.1 ± 1.4
Periclase     6.10 ± 0.38 4.41 ± 0.24 3.3 ± 1.9

Note: Each of C3S, C2S, C3A, and C4AF are known with several phases with the same composition but slightly different structures.

The AI-powered phase identification’s quantitative analysis results are closer to the certified values than those of the conventional search/matches, indicating their higher accuracy compared to the traditional search/match.

How does AI help me?

This AI-powered module can suggest how to adjust your simulation model to improve the quality and accuracy of X-ray reflectivity (XRR) analysis.

A trained network can suggest what might be causing the discrepancy between the experimental and simulated profiles, such as a missing surface layer. Correcting the simulation model based on the suggestion can improve the accuracy of the XRR fitting.

How does AI learn to predict the cause of a discrepancy?

This module uses Vision Transformer, a neural network-based deep-learning algorithm. A neural network can be trained and learn what type of discrepancy between measured and simulated XRR profiles appears for the particular error in the model by using simulated data.

AI schematic

How does a neural network learn?

The neural network is modeled loosely on the human brain. The network architecture has input, hidden, and output layers with a number of “neurons.” The neurons are connected between the layers, and random weights and biases are assigned to these connections to set their “importance.” During training, the network processes input data (simulated and experimental XRR patterns) and predicts the output (possible error) using the initial weights and biases. Then, the prediction is compared to the ground truth. Depending on how close the output is to the ground truth, a “reward” or “penalty” is propagated through the network as “feedback.” By repeating this process, the network refines the weights and biases of the connections, which is equivalent to “learning” the relationship between the input and output.

Example 1

The simulation (blue line) is based on a model with two layers on a substrate. It does not reproduce the experimental profile (red line) well. The AI-assist model suggested the thickness of the first and second layers are reversed.

Example 1 graphic

 

Example 2

The simulation (blue line) is based on a model with a superlattice on a substrate. It does not reproduce the experimental profile (red line) well. The number of fringes between the superlattice peaks does not match between the simulation and the experiment. The AI-assist model suggested the number of the superlattice period is incorrect.

Example 1 graphic

Performance

Seven and nine types of suggestions are available for 1-2 layer models and superlattice models, respectively. The suggestion has been correct nine out of ten times in the cases tested so far.

Example 3

The simulation (bluevline) is based on a model with a superlattice on a substrate. It does not reproduce the experimental profile (red line) well. The superlattice peaks stay narrow in the simulation, while the experimental data shows a broadening of the same peaks. The AI-assist model suggested the thickness (period) of the superlattice is not uniform.

Example 1 graphic

Example 4

The simulation (blue line) is based on a model with two layers on a substrate. It does not reproduce the experimental profile (red line) well. The AI-assist model suggested a thin layer at the top surface is missing.

Example 1 graphic

Adding the top surface layer to the model significantly improved the XRR fitting.

Example 1 graphic

How does AI help me?

This AI-powered module can separate an X-ray diffraction (XRD) pattern of an unknown mixture into multiple components and quantify each phase.

This decomposition does not require any knowledge or phase identification analysis of the sample and can be useful when comparing tens and hundreds of XRD patterns consisting of similar phases.

How does AI-powered XRD profile decomposition work?

This module applies a demensionality-reduction algorithm originally developed for machine learning. The dimensionality-reduction algorithm attempts to reproduce an XRD pattern by combining a few XRD patterns at different weights.

It starts off with randomly guessed phases and gradually increases the number of phases as necessary while refining the weights and trying to fit the experimental XRD pattern each time. It repeats the process, trying thousands of combinations, until the experimental XRD profile is reproduced sufficiently well by a combination of a few components. A user can specify the number of phases or let the algorithm estimate it.

The components can be any crystalline material, including unknown phases, or even amorphous.

When is AI-powered XRD profile decomposition useful?

AI-powered XRD profile decomposition is beneficial when you need to analyze many similar XRD patterns.

For example, in-situ measurements, especially DSC-XRD and time-resolved measurements, can produce hundreds of XRD patterns to analyze. Measurements of a 96-well plate, often used for combinatorial chemistry, can also produce many patterns. When analyzing these data, using the AI-powered profile decomposition can produce phase identification and quantitative analysis results with minimum operator involvement.

 

Example

An example of similar diffraction patterns is shown below.

Example 1 graphic

The profile decomposition algorithm suggested three phases, including an amorphous phase.

Example 1 graphic

In the decomposition process, each phase is quantified based on its weight.

Example 1 graphic